TECHNICAL REPORTS Atmospheric Pollutants and ... - PubAg - USDA

5 downloads 0 Views 619KB Size Report
TECHNICAL REPORTS. Atmospheric Pollutants and Trace Gases. Ammonia Emissions from Swine Houses in the Southeastern United States. Lowry A. Harper ...
Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

TECHNICAL REPORTS Atmospheric Pollutants and Trace Gases Ammonia Emissions from Swine Houses in the Southeastern United States Lowry A. Harper,* Ron R. Sharpe, and John D. Simmons ABSTRACT

L.A. Harper and R.R. Sharpe, Southern Piedmont Conservation Research Unit, USDA-ARS-JPCSNRCC, 1420 Experiment Station Road, Watkinsville, GA 30677. J.D. Simmons, Poultry Research Unit, USDA-ARS, 606 Spring Street, Starkville, MS 39759. Contribution from the USDA-ARS-JPCNRCC, USDA-ARS Poultry Research Unit, and University of Georgia Agricultural Experiment Station. Received 30 Dec. 2002. *Corresponding author ([email protected]).

son and Grennfelt, 1988) and when excessive quantities of nitrogen (N) are deposited onto the landscape, nitrophilous species become better competitors. However, in cropping systems, atmospheric NH3 and NH4⫹ may be beneficial by adding N during critical times of the day (Harper et al., 1987) and during periods of soil N deficiency (Sharpe et al., 1988; Harper et al., 1996). Crop canopies may also remove significant quantities of NH3 released to the atmosphere from nearby sources (Harper and Sharpe, 1995; Harper et al., 1996; Bussink et al., 1996; Sharpe and Harper, 2002). Concentrated livestock production can be a significant source of NH3 emissions to the atmosphere in a relatively small geographic area. Adverse effects may be due to the direct and indirect effects of NH3 (which has a shorter residence time of hours) and/or NH4⫹ aerosols of nitrate (NO3⫺) and sulfate (SO42⫺), which may have a residence time of 5 to 9 d (Crutzen, 1983). Early estimates (Hatfield et al., 1993) suggested that 89 to 90% of the N input to anaerobic lagoons is lost to the atmosphere through NH3 volatilization. These estimates represent about 60% of the total feed N input into the farm operation. Current estimates in North Carolina (Doorn et al., 2002) suggest that 36% of the total N going into confined animal feeding operations (CAFOs) in the state is volatilized as NH3 gas from all sources including lagoons, houses, and field applications. However, other studies in the North Carolina and Georgia Coastal Plains region of the USA (Harper and Sharpe, 1998; Harper et al., 2004) have shown that lagoons emit significantly less NH3 than previously thought. Harper et al. (2004) found in a highly measured swine production operation that about 5% of the N going into the operation as feed left the lagoon as volatile NH3 and another 1% from field application of waste effluent. Much of the N (about 43% of input feed) that entered into lagoons was found to be denitrified to N2 (Harper and Sharpe, 1998; Harper et al., 2000, 2004) by microbial and/or chemical (Van Cleemput, 1998) denitrification. Another source of NH3 emissions that has not been comprehensively measured is emissions from animal production houses. The purpose of this research was to evaluate NH3 swine confinement housing emissions and

Published in J. Environ. Qual. 33:449–457 (2004).  ASA, CSSA, SSSA 677 S. Segoe Rd., Madison, WI 53711 USA

Abbreviations: AU, animal unit (1 AU ⫽ 500 kg); DOY, day of year; FF, farrow-to-finish; FW, farrow-to-wean.

Ammonia (NH3) from confined animal feeding operations is emitted from several sources including lagoons, field applications, and houses. This paper presents studies that were conducted to evaluate NH3 emissions from swine finisher and sow animal houses in the southeastern USA. Management and climate variables including animal weight, feed consumption, housing gutter water temperature, total time fans operated per day, house air temperature, house ambient NH3 concentration, and animal numbers were measured to determine their individual and combined effect on NH3 emissions. Ammonia emissions varied on daily and seasonal bases with higher emissions during warmer periods. For finishers, the summertime housing emissions on a per-animal basis were 2.4 times higher than wintertime (7.0 vs. 3.3 g NH3 animal⫺1 d⫺1) or 3.2 times higher when compared on an animal unit (AU) basis (1 AU ⫽ 500 kg) because of climate and animal size differences between measurement periods. For summertime, the emission factor for the finishing pigs was 7.8 times higher than for sows on an animal basis and 25.6 times higher on an AU basis. Simple models were developed for housing emissions based on (i) all measured factors that were independent of each other and (ii) on three commonly measured management factors. The two models explained 97 and 64%, respectively, of variations in emissions. Ammonia emissions were found to be somewhat less than other studies on the same type housing due to more representative housing concentration measurements and calibration of exhaust fans; thus, emission factors for these type houses will be less than previously thought.

A

mmonia is ubiquitous and is the major atmospheric alkaline component that neutralizes acid gases produced by burning fossil fuels. The neutralization process produces ammonium (NH4⫹), which is a major component of atmospheric aerosols (particulate matter) and rainfall (Asman, 1994). Wet and dry deposition, whether from agriculture, industry, or transportation, may exacerbate soil acidification (van der Molen et al., 1990) and possible plant nutrient imbalances in natural ecosystems. Additionally, many natural systems such as forest and heath are adapted to low nutrient conditions (Nils-

449

450

J. ENVIRON. QUAL., VOL. 33, MARCH–APRIL 2004

Table 1. Management characteristics for the swine production units. Measurement period

Farm

Management type

Animal number

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

house⫺1

11–16 February 2–6 February 22–28 July 28–31 July 8–18 August

10 20 10 10 20

farrow farrow farrow farrow farrow

to to to to to

finish wean finish (#9) finish (#8) wean

determine the amount of N as a fraction of feed input that leaves the farm system. MATERIALS AND METHODS The swine production houses described in this paper were “farrow-to-finish” (FF) and “farrow-to-wean” (FW) units at different locations in the Coastal Plains of North Carolina. The FF farm contained about 1200 sows and boars, 1400 nursing pigs, and 7500 finishers. The finishers, pigs being grown for market, ranged in size from 20 to 120 kg. Both of the houses measured at the FF farm contained finishing pigs only. The FW farm contained sows, boars, and nursing pigs. The house measured at the FW farm contained 886 sows. Table 1 presents management and facility characteristics for the houses. The FF houses were “flush”-type with recycled water from the lagoon used for flushing the gutters beneath the slatted floors. Effluent from the lagoon was pumped to flush tanks and every four hours two of the four gutters were flushed. The gutters had a 0.5% slope representing a 0.30- to 0.41-m water depth from entrance to exit to remove manure and urine during the flush cycle. The water depth was 0.30 m from the slat flooring. Four hours later the other two were flushed. The FW house used a “pull-plug,” pit-recharge system beneath the slatted floors with a cycle time of about one week between drain and recharge. Two houses were instrumented at the FF farm with sensors to determine cycling of the individual forced-ventilation fans. Closed-path, tunable diode laser spectrometers (Edwards et al., 1994; Dias et al., 1996) (Trace Gas Analyzer System; Campbell Scientific, Logan, UT) were used to measure NH3 and methane (CH4) concentration differences between the intake and exhaust points of the houses. Measurements of CH4 emissions were previously reported by Sharpe et al. (2001). Inside NH3 concentrations were measured about 10 m from the exhaust fans to avoid variable air trajectories under different fan-on conditions. Ammonia concentrations were measured 10 times per second and differential concentrations (differences between the incoming and exiting air masses) were calculated every minute. Ambient temperature inside the houses and water temperature of the waste pits were measured with recording temperature sensors (WTA08; Onset Computer Corp., Bourne, MA). A portable fan airflow measurement system was mounted to each fan to measure total volumetric flow rate of the ventilation fans (Simmons et al., 1999) for fan efficiency evaluation. Fan flow rates are affected by maintenance, age of the fan, and the interference of flow rates by other fans nearby [reduced air pressure and interference of flow trajectories (Simmons et al., 1999)]. Each fan was measured to determine its efficiency, which varied slightly from the rated capacity due to aging and the number of fans operating. Fan efficiencies were measured for all the fans in each building independently and in combination sequence, which might occur due to the programmed sequence of all other fans. The fans ranged in efficiency from 82 to 98% depending upon the individual fan and the number and sequence of fans on at any time. Fan efficiencies were then applied to fans during the studies. Individual cup anemometers were used to determine when individual fans were operating.

animals 779 886 873 904 884

Animal average weight kg

animal⫺1 90.8 208.8 56.8 62.6 208.8

Feed consumption kg animal⫺1 d⫺1 2.2 2.2 1.6 1.4 1.9

Two seasons of NH3 emissions measurements were made during the annual climatic extremes, winter (11–16 Feb. 1998) and summer (22–31 July 1998) for the FF houses and 8 to 18 Aug. 1998 for the FW house. A schedule of five to eight days of 24-h data collection periods were made throughout the winter and summer measurement seasons. Emissions were calculated from the trace-gas differentials of incoming and exhausting air and the exhausted air volumes produced by the fans. Fans were controlled by timing and/or temperature sensors to (i) maintain acceptable concentrations of NH3 in the houses and (ii) maintain acceptable temperature levels. The NH3 control was the dominant control with intermittent cycling (fan on approximately 8.5 min and off approximately 3.5 min) and inside temperatures were controlled by increasing numbers and sizes of fans operating. The equation used to calculate NH3 flux was:

NH3 emissions (g min⫺1) ⫽ ⌬NH3 (g m⫺3) ⫻ fan capacity (m3 min⫺1) ⫻ fan efficiency

[1]

The sum of NH3 emission rates from each fan equaled total flux from the house. Calculations of NH3 flux from the houses are based on the assumption that all emissions are through the exhaust fans at the end of the houses. This assumption may result in an underestimation of total flux during the winter period when fans were intermittently on. However, the underestimation is probably minimal because even during winter, fans operated a specified percentage of time to maintain NH3 concentrations in the houses at an acceptable concentration. The FF houses had a set of five fans in the east end of the house. Fans 1, 3, and 5 had a rated capacity of 650 m3 air min⫺1 and fans 2 and 4 had a capacity of 310 m3 air min⫺1. The FW house had six fans with rated capacities of 650 m3 air min⫺1.

RESULTS AND DISCUSSION Differential NH3 concentrations between the entrance (local ambient concentration) and the fan exit varied between seasons and production types. Figure 1 gives a trace of fan activity for a FF house and NH3 differential concentrations between outside and inside the house. Outside ambient concentrations varied between 0.1 and 2 ␮g NH3 g⫺1 depending on if the wind direction was from the forested areas nearby or from the lagoons or other buildings. Measurements from Day of Year (DOY) 28.9 to 29.0 in Fig. 1 showed the simplest type response in concentration variation to fan activity with only one fan cycling. The variation in NH3 differential showed that when a fan turned on, the NH3 concentration decreased quickly, but then gradually increased to a concentration approaching that of when the fans were off. We think that when a fan (or fans) was actuated, air initially was exchanged only from the house interior, whereas when the fan remained on for an extended period, air from the gutters mixed into the inte-

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

HARPER ET AL: AMMONIA EMISSIONS FROM SWINE HOUSES

451

Fig. 1. The effect of wintertime fan activity on inside–outside house NH3 differential concentrations. Fan status (on–off) was determined by measuring windspeed through the fans. The variability in lines at the off status are indicative of outside turbulence affecting the anemometers.

rior of the house increasing the interior concentration. Gutter fans in many types of FW houses (for smaller animals) are used to avoid mixing gutter gases into the house interior. Increased fan activity (summertime conditions, Fig. 2) did result in decreased house NH3 concentrations with a seasonal decrease from 7.0 ␮g NH3 kg⫺1 in winter to 3.3 ␮g NH3 kg⫺1 in summer. Insidehouse ambient NH3 concentrations during wintertime and summertime were similar in magnitude (Fig. 2 and 3) when the fans were off in winter and the minimum number of fans on in summer. However, when the fans (or more fans) cycled on, the concentrations dropped quickly. Under summertime conditions the inside– outside differential can be seen to be generally a function of when the most fans (and their respective sizes) are on or off (Fig. 2). Gutter water temperature during the measurement periods in winter ranged between 14 and 16⬚C due to the heat capacity of the building and soil surrounding the gutter, except for a short period after the gutter was flushed with lagoon effluent. The gutter temperature, measured 0.1 m above the gutter floor in the effluent, decreased about 4⬚C in winter for about 45 min. In summer, the gutter temperature increased about 5 to 7⬚C because of the warm lagoon effluent flush water. Measured finishing housing emissions during the winter ranged from no emissions (when all fans were off) to a short-term maximum of about 7.5 g NH3 min⫺1 house⫺1 (Fig. 3). The average seasonal winter emission was 2.57 kg NH3 d⫺1 house⫺1 (3.3 g NH3 animal⫺1 d⫺1

or 18.1 g NH3 AU⫺1 d⫺1) (an animal unit, AU, is based on an animal weight of 500 kg) or 5.7% of feed N input to the farm. Summer finishing house emissions ranged from a minimum of 3 kg NH3 d⫺1 house⫺1 (since fans were on continuously for temperature control) to as high as 14 kg NH3 d⫺1 (Fig. 2) giving a summer seasonal average of 6.25 kg NH3 d⫺1 house⫺1 (7.0 g NH3 animal⫺1 d⫺1 or 58.9 g NH3 AU⫺1 d⫺1) or 8.2% of feed N input to the farm. Ammonia emissions are generally a function of the physical and chemical parameters of solution NH4⫹ concentration, pH, and temperature, and air turbulence to remove NH3 from the water–air boundary layer. Summertime housing emissions were larger (2.4 times) than wintertime, due to higher gutter water temperature and inside turbulence and air exchange. There was no emissions effect due to winter and summer lagoon NH4⫹ and pH (used for flush water) since they were not significantly different between seasons (Harper et al., 2004). Gutter NH4⫹ and pH were not measured because the houses were flushed every four hours. On a per-AU basis, NH3 emissions were 3.2 times higher in summer than winter due to climate and animal size differences for the respective seasons. Annual emissions for the FF farm were 14 347.4 kg N yr⫺1 representing 7.3% of feed input to the farm. Summertime sow house emissions ranged from no emissions to a maximum of 3.5 g NH3 min⫺1 house⫺1. The seasonal average summertime emission was 0.80 kg NH3 d⫺1 house⫺1 (0.9 g NH3 animal⫺1 d⫺1 or 2.09 kg NH3 AU⫺1 d⫺1) or 1.8% of feed N input to the farm.

J. ENVIRON. QUAL., VOL. 33, MARCH–APRIL 2004

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

452

Fig. 2. Summertime 24-h NH3 emissions and concentrations in a swine finishing house in response to climate and fan activity. Fan status (on–off) was determined by measuring windspeed through the fans. Differences in the magnitude of the fan windspeeds result from different fan efficiencies and fan sizes.

453

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

HARPER ET AL: AMMONIA EMISSIONS FROM SWINE HOUSES

Fig. 3. Wintertime 24-h NH3 emissions and concentrations in a swine finishing house in response to climate and fan activity. Fan status (on–off) was determined by measuring windspeed through the fans. Differences in the magnitude of the fan windspeeds result from different fan efficiencies and fan sizes.

454

J. ENVIRON. QUAL., VOL. 33, MARCH–APRIL 2004

Table 2. Comparison of annualized emission factors on a per-animal basis for finishing operations. Location

Emission factor

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

kg NH3 U.S. Midwest North Carolina, North Carolina, North Carolina, North Carolina, North Carolina,

Farm Farm Farm Farm Farm

10 10 10 10 10

animal⫺1 4.68 4.81 3.36 2.57 3.05 1.89

Type

Reference

house ⫹ gutter fans, summer only daylight measurement only, summer only daylight measurement only, summer only total daily measurement, summer only daylight measurement only, annual total daily measurement, annual

Parbst et al. (2000) Harris et al. (2001) Harris and Thompson (1998) this study Harris and Thompson (1998) this study

yr⫺1

Summertime emissions of the finishing houses were 7.8 times higher than for sows and 25.6 times higher on an AU basis. Wintertime emissions are not available for the sows, but an annual estimate of the FW house emissions, based on summertime emissions, was less than 1% of feed N emitted to the atmosphere as NH3. Studies by Harris and Thompson (1998) on the same houses suggested little seasonal variation existed between winter and summer NH3 emission rates (7.5 g NH3 animal⫺1 d⫺1 in winter and 9.2 g NH3 animal⫺1 d⫺1 in summer) because the animals were kept in a reasonably constant environment with little stress from outside environmental factors. The differences between studies are probably due to the fact that the measurements of Harris and Thompson (1998) were determined during daylight periods and extrapolated to an entire day. Also, because their concentrations were measured outside about 10 m from the exhaust fans it is possible that some of the plumes may have been entrained to the inlets, providing a significant increase in background or incoming NH3 concentration. Comparison of summertime emissions on an annualized basis from several studies show considerable differences in emissions (Table 2). Although the study of Harris and Thompson (1998) and this study were on the same farm, the differences may be attributable to measurement period and location. The difference between the results of Parbst et al. (2000) and the current study may be due to the use of gutter fans (with an associated increased turbulence in the pit area) to minimize house NH3 concentrations in the Parbst et al. (2000) study. Table 2 also exemplifies the difference between the use of annualized data taken only in one season compared with annual measurements since ambient climatic conditions will affect emissions even though the housing climate is regulated. Seasonal differences between animal house types are evident in Fig. 2 and 3. Ammonia concentrations in the houses were slightly higher during winter than summer because of decreased air exchange for heat conservation. During winter (Fig. 3) a slight daily variation in house concentrations and emissions occurred due to the duration of time the primary fan, which was programmed to cycle for NH3 removal, was on. As long as the cycling was constant (e.g., DOY 45.8–46.3), inside NH3 concentrations and emissions followed the fan cycling. However, when inside ambient temperature required the fan to stay on for longer periods, the NH3 concentrations increased (except DOY 45.6) with a slight increase in emissions. Similarly, when the primary fan remained on for most of the time or a secondary fan came on, NH3 concentrations increased significantly

(e.g., DOY 44.3–44.7) along with an increase in emissions. It is interesting that when the primary fan remained on for longer periods, the house concentrations did not drop similar to the periods when the fan was on intermittently. We think that when the fan remained on for longer periods, turbulence structure in the house was larger and pulled NH3 from the pit area. This phenomenon was observed throughout the winter season measurements. Many types of swine houses have gutter fans to reduce NH3 concentrations inside the house, particularly when the animals are small. Initiation of a secondary fan increased emissions considerably (e.g., DOY 44.6). During summer, there were some fans on continuously and the house concentrations were generally a function of the number of fans operating. There were increased NH3 concentrations and emissions on a daily basis around sunup each day. Fan activity did not correlate with the spikes and we think the increases were due to animal activity (wake-up and/or feeding times). Ammonia emissions of sows and finishers were quite different (Fig. 2 and 4). Concentrations during summer for the sows were much lower than finishers. During nighttime, the inside–outside concentration differentials for the sows approached zero and emissions were very small. The emission factor for finishers was 7.8 times higher than sows. Seasonal average emissions of sows were 0.80 kg NH3 d⫺1 house⫺1 (0.9 g NH3 animal⫺1 d⫺1 or 2.30 kg NH3 AU⫺1 yr⫺1). There was a distinct daily variation in the sow house similar to the finishers with increases in concentrations and emissions beginning about sunup and feeding time. We think that the much greater concentrations in the house during daytime were due also to increased turbulence when all the fans were running. Table 3 presents average daily information on measurements and input production data for the finisher and sow houses for the two seasons of measurement. Due to equipment malfunction, winter data for the sow houses are not available. There were variations in seasonal daily averages due mainly to change in ambient microclimate. Multiple regression analysis suggested that animal size, total duration when fans were operating, and NH4⫹ content of the input flush water were the dominant factors affecting house emissions. A regression model (number of observations ⫽ 21 daily average values over the summer and winter periods, df ⫽ 16) estimating individual housing emissions based on these management and measured factors explained 97% of the variability in emissions: FNH3 ⫽ (⫺0.6955 ⫻ AW) ⫹ (4.42 ⫻ 10⫺5 ⫻ tf) ⫺ (0.1923 ⫻ NH⫹ 4 ) ⫹ (70.0802 ⫻ Cf) ⫹ (0.7931 ⫻ Tgw) [2]

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

HARPER ET AL: AMMONIA EMISSIONS FROM SWINE HOUSES

455

Fig. 4. Summertime diurnal NH3 emissions and concentrations in a swine sow house in response to climate and fan activity. Fan status (on–off) was determined by measuring windspeed through the fans. Differences in the magnitude of the fan windspeeds result from different fan efficiencies and fan sizes.

where FNH3 is the NH3 housing emission rate in kg NH3 d⫺1, AW is the average animal weight in kg animal⫺1, tf is the total time all fans were operating per day (min d⫺1 for all fans), NH4⫹ is the ammonium content of input flush water in ␮g g⫺1, Cf is feed consumption in kg animal⫺1 d⫺1, and Tgw is the gutter water temperature (⬚C). The valid ranges of input values for the statistical relationships are AW ⫽ 90 to 300 kg animal⫺1, tf ⫽ 650 to 15 000 min d⫺1 for all fans, NH4⫹ ⫽ 160 to 550 ␮g

g⫺1, Cf ⫽ 1.5 to 2.2 kg animal⫺1 d⫺1, and Tgw ⫽ 15 to 29⬚C. Individual coefficient standard errors for AW, tf, NH4⫹, Cf, and Tgw were 0.0761, 6.46 ⫻ 10⫺5, 0.0222, 7.9841, and 0.0650, respectively. A regression of common production and easily measured variables explained 64% of the variability in emissions. This simple predictive relationship, including input data for animal weight, gutter washwater NH4⫹ content, and feed consumption, is shown in Eq. [3]:

456

J. ENVIRON. QUAL., VOL. 33, MARCH–APRIL 2004

Table 3. Average daily information from finisher and sows houses in North Carolina.

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

Animal type Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Finishers Sows Sows Sows Sows Sows Sows Sows Sows

Day of year

Ammonia emissions

House ammonia concentration

Total time all fans on

House air temperature

28 38 44 45 46 205 206 207 208 209 210 211 212 221 222 223 225 226 227 228 229

kg NH3 d⫺1 1.654 3.401 3.310 2.975 1.489 8.600 7.460 6.819 6.972 6.937 7.443 2.967 2.819 0.636 0.660 0.598 1.312 0.324 1.054 1.226 0.586

␮g NH3 g⫺1 14.4 12.4 8.7 8.4 5.5 3.3 3.7 3.8 4.3 4.0 3.6 1.7 1.9 0.5 0.5 0.5 0.5 0.1 0.1 0.1 0.1

min d⫺1 1 039 636 1 168 1 063 829 6 010 4 632 4 401 3 890 3 689 4 641 5 879 5 883 14 400 12 872 14 277 14 400 10 706 9 124 9 124 5 568

23.1 23.1 23.2 22.9 23.1 30.5 28.4 28.0 27.4 27.4 27.4 27.3 27.3 26.2 25.5 24.9 24.9 25.0 25.5 26.1 26.3

ⴗC

FNH3 ⫽ (0.2065 ⫻ AW) ⫹ (0.0723 ⫻ NH4⫹) ⫺ (24.3307 ⫻ Cf)

Gutter water temperature 15.1 15.1 15.1 14.9 14.9 29.0 28.4 28.2 26.9 28.2 27.7 27.4 27.8 26.3 26.2 24.9 24.9 25.0 25.5 26.1 26.3

Animal numbers

Animal weight

Feed consumption

animals house⫺1 779 779 779 779 779 873 873 873 873 873 873 904 904 884 884 884 884 884 884 884 884

kg animal⫺1 90.8 90.8 90.8 90.8 90.8 56.8 56.8 56.8 56.8 56.8 56.8 62.6 62.6 196.6 196.6 196.6 196.6 196.6 196.6 196.6 196.6

kg animal⫺1 d⫺1 2.24 2.24 2.24 2.24 2.24 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.91 1.91 1.91 1.91 1.91 1.91 1.91 1.91

REFERENCES [3]

where number of observations ⫽ 21, df ⫽ 18, and the standard errors of the coefficients are 0.0520, 0.0156, and 6.0649, respectively.

CONCLUSIONS Ammonia emissions were measured from finishing and sow housing under summertime and wintertime conditions. Emissions were found to be somewhat less than other studies on the same type housing due to total daily measurement, possibly more nearly representative inside housing NH3 concentrations, actual calibration of fans, and totalization of individual fan operation during the measurement. Emissions from this type of animal house were also found to be somewhat less than similar emissions measurements made in the U.S. Midwest. However, this type of housing did not have pit fans to remove NH3 from the gutters, a factor that would increase the turbulence in the gutter area causing higher emission rates for the same type animals. Regression models were developed to predict emissions from the houses and if all measured management and environmental independent factors were included in the model, 97% of the variability in emissions was explained. A simple regression model of commonly measured production factors explained 64% of the variability in emissions. ACKNOWLEDGMENTS These studies were supported, in part, by the USDA-ARS Global Change National Program, the North Carolina Pork Council, and the individual farm owners. The authors would like to acknowledge the technical assistance of J.E. Scarbrough, M.A. Thorton, S. Norris, P. Caddis, M.G. Hunter, and J. Merritt in these studies. Thanks are expressed for editorial suggestions by G.W. Boyd, N.A. Cole, D.B. Harris, R. Todd, and K. Weaver.

Asman, W.A.H. 1994. Emission and deposition of ammonia and ammonium. Nova Acta Leopold. 70:263–297. Bussink, D.W., L.A. Harper, and W.J. Corre. 1996. Ammonia transport in a temperate grassland: II. Diurnal fluctuations in response to weather and management conditions. Agron. J. 88:621–626. Crutzen, J. 1983. Atmospheric interactions—Homogeneous gas reactions of C, N, and S containing compounds. p. 67–112. In G. Bolin and R.B. Cook (ed.) The major biochemical cycle and their interactions. Wiley, Chichester, UK. Dias, G.T., G.W. Thurtell, C. Wagner-Riddle, and L.A. Harper. 1996. Measuring ammonia fluxes from soil with a laser-based trace gas analyzer. In Proc. Int. Conf. on Air Pollut. from Agric. Operations, Kansas City, MO. 7–9 Feb. 1996. Am. Soc. Agric. Eng., St. Joseph, MI. Doorn, M.R.J., D.F. Natschke, and P.C. Meeuwissen. 2002. Review of emission factors and methodologies to estimate ammonia emissions from animal waste handling. Rep. EPA-600/R-02-017. USEPA, Washington, DC. Edwards, G.C., H.H. Neumann, G. den Hartog, G.W. Thurtell, and G.E. Kidd. 1994. Eddy correlation measurements of methane fluxes using a tunable diode laser at the Kinosheo Lake tower site during the Northern Wetlands Study (NOWES). J. Geophys. Res. [Atmos.] 99:1511–1517. Harper, L.A., D.W. Bussink, H.G. van der Meer, and W.J. Corre. 1996. Ammonia transport in a temperate grassland: I. Seasonal transport in relation to soil fertility and crop management. Agron. J. 88:614–621. Harper, L.A., and R.R. Sharpe. 1995. Nitrogen dynamics in irrigated corn: Soil-plant nitrogen and atmospheric ammonia transport. Agron. J. 87:669–675. Harper, L.A., and R.R. Sharpe. 1998. Atmospheric ammonia: Issues on transport and nitrogen isotope measurement. Atmos. Environ. 32:273–277. Harper, L.A., R.R. Sharpe, T.B. Parkin, A. De Visscher, O. van Cleemput, and F.M. Byers. 2004. Nitrogen cycling through swine production systems: Ammonia, dinitrogen, and nitrous oxide emissions. J. Environ. Qual. (in press). Harper, L.A., R.R. Sharpe, G.W. Langdale, and J.E. Giddens. 1987. Nitrogen cycling in a wheat crop: Soil, plant, and aerial nitrogen transport. Agron. J. 79:965–973. Harper, L.A., R.R. Sharpe, and W.P. Robarge. 2000. Ammonia and other nitrogen emissions from swine waste lagoons in the U.S. Southeastern Coastal Plains. p. 191–203. In Proc. Workshop on Atmospheric Nitrogen Compounds II: Emissions, Transport Transformation, Deposition, and Assessment, Chapel Hill, NC. 7–9 June 1999. North Carolina Dep. of Environ. and Nat. Resour., Raleigh. Harris, D.B., R.C. Shores, and L.G. Jones. 2001. Ammonia emission

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

HARPER ET AL: AMMONIA EMISSIONS FROM SWINE HOUSES

factors from swine finishing operations. In Int. Emissions Inventory Conf.: One Atmosphere, One Inventory, Many Challenges [Online]. Available at http://www.epa.gov/ttn/chief/conference/ei10/ ammonia/harris.pdf (verified 5 Dec. 2003). USEPA, Washington, DC. Harris, D.B., and E.L. Thompson, Jr. 1998. Evaluation of ammonia emissions from swine operations in North Carolina. p. 420–429. In Proc. Emission Inventory—Living in a Global Environment, VIP88, New Orleans, LA. 8–10 Dec. 1998. Air & Waste Manage. Assoc., Pittsburgh, PA. Hatfield, J.C., M.C. Brumm, and S.W. Melvin. 1993. Swine manure management. p. 40–57. In Agricultural utilization of municipal, animal, and industrial waste. Rep. CRR 44. USDA Agric. Res. Serv., Washington, DC. Nilsson, J., and P. Grennfelt. 1988. Critical loads for sulphur and nitrogen. Rep. 15. Nordic Council of Ministers, Copenhagen. Parbst, K.E., K.M. Keener, A.J. Heber, and J.Q. Ni. 2000. Comparison between low-end discreet and high-end continuous measurements of air quality in swine buildings. Appl. Eng. Agric. 16:293–299.

457

Sharpe, R.R., and L.A. Harper. 2002. Nitrous oxide and ammonia emissions from a soybean field fertilized with swine effluent from a waste holding lagoon. J. Environ. Qual. 31:524–532. Sharpe, R.R., L.A. Harper, J.E. Giddens, and G.W. Langdale. 1988. Nitrogen use efficiency and nitrogen budget for conservation tilled wheat. Soil Sci. Soc. Am. J. 52:1394–1398. Sharpe, R.R., L.A. Harper, and J.D. Simmons. 2001. Methane emissions from swine houses in North Carolina. Chemosphere Global Change Sci. 3:1–6. Simmons, J.D., T.E. Hannigan, and B.D. Lott. 1999. A portable anemometer to determine the output of large in-place ventilation fans. Appl. Eng. Agric. 14:649–653. Van Cleemput, O. 1998. Subsoils: Chemo- and biological denitrification, N2O and N2 emissions. Nutr. Cycling Agroecosyst. 52:187–194. Van der Molen, J., A.C.M. Beljaars, W.J. Chardon, W.A. Jury, and H.G. van Faassen. 1990. Ammonia volatilization from arable land after application of cattle slurry. 2. Derivation of a transfer model. Neth. J. Agric. Sci. 38:239–254.